(Revised) DESCRIPTION (provided by applicant): The proposed study will examine the impact of administered alcohol on women's recognition of aggression risk and responses to aggression using experimental analog methodology. Although many studies have shown an association between women's alcohol consumption and victimization, few have examined the direct effects of alcohol on women's vulnerability or explored the mechanisms by which increased vulnerability results. The study promises to contribute to theoretical understanding of alcohol and victimization, while providing practical guidance in developing effective prevention programs for women. This R21 proposal addresses NIAAA's special emphasis area on the relationships between alcohol and violence. It is an outgrowth of Dr. Testa's currently funded Independent Scientist Award and Women 2000 project. Participants will consist of 105 unmarried women, 21-32 years of age, who were recruited from the community using random digit dialing, completed participation in the Women 2000 project, and meet eligibility criteria. Women will be randomly assigned to receive alcohol (sufficient to raise BAL to .08), no alcohol, or a placebo. They will respond to a 3-part written vignette describing a hypothetical situation. After Part 1, in which the situation is described positively, perceptions of the man and intended behaviors will be assessed. The second part of the scenario includes cues suggestive of aggression, while the third part depicts more obvious aggression. Risk recognition and anticipated responses to aggression will be assessed after Parts 2 and 3. Compared to sober women, women who consume alcohol are expected to perceive less risk, anticipate engaging in more risky behavior, and anticipate less resistance to the man's aggressive behavior. However, consistent with alcohol myopia theory, we predict that alcohol-related impairment will be relatively greater when aggression cues are subtle compared to the situation in which aggression is salient. The role of several potential covariates: liking for man, assertiveness, alcohol expectancies, and previous aggression, also will be considered.